solar power for security alarm system

[Sunking]

It may be a few days, as we have a lot of projects going on right now that need my attention.

If you have the luxury of time and a Kill-A-Watt meter on your side, take up to a week to get good data. For example run it for exactly 1-week say starting Wednesday just at Sunrise, and then run it exactly 7 days. Take that result and divide by 7. That will give very accurate data. Don't worry about what the siren will use as it is insignificant. Example if the siren uses 100 watts for 2 minutes is only 3 to 4 watt hours.

I have a little experience with alarm systems, mostly fire, and the power use if memory serves me correctly is only in the 15 to 20 watt range. Si would guess you will the alarm system uses about 400 to 500 watt hours per day which is doable. Lets us know what you find out. Only takes about 30 seconds to do a design.

One warning here though you need to pay attention to is the physical location because it posses a few challenges. Th epanels need to be located with unobstructed view of the horizon East-West and South. In addition the charge controller, batteries, and inverter need to be very close to the panels.

[fraggedout]

It is a good and fair question that gets asked a lot, perhaps I should post a canned answer for a Sticky sometime.

Please do, as I'm sure hammering out a response every time someone asks this question gets a bit cumbersome and repetitive. Thank you for taking the time to reply with such a detailed and satisfactory explanation. I'll do some testing and come back to the forum when I have more concrete data to work with. It may be a few days, as we have a lot of projects going on right now that need my attention.

[Sunking]

Please point out the flaw(s) in my logic.

It is a good and fair question that gets asked a lot, perhaps I should post a canned answer for a Sticky sometime.

Short story is there is more that can go wrong with parallel arrangements than serial arrangements. However I need to go over a few problems.

With a parallel arrangement you cannot equalize load or charge currents for a few reasons:

Internal battery resistance varies.
Wire lengths vary.
Connection resistance and workmanship varies.

What this does is cause one string in the parallel to do most of the work and causes significant wear and tear on the string, until the point it forces the other string to pick up the slack and cause premature failure on the second string. Most battery failures, not all result as high resistance or open circuit. For average Joe Home owner an open circuit is impossible to detect until both string have failed. The only way to detect it is with a hydrometer or well documented battery logs keeping track of individual cell voltages. I will bet you money in your biz you use 12 volt sealed batteries in which you cannot even take Specific Gravity or individual cell voltage measurements.

Second problem which is less frequent, but a real problem with sealed batteries is a shorted out cell. In a series arrangement is not much of a problem, as both load and charge currents easily pass through a shorted cell. It is also very easy to detect by the voltage check of loosing 2 volts, but still operational. However in a parallel arrangement, a shorted cell in one string is a major problem likely to cause a fire.

In a series arrangment load and charge currents are always equal no matter what. The worse thing that can happen and it will happen is a cell with go high resistance or open circuit like all batteries do.

So series affords the longest possible battery life, safer, and the lessor of the two evils.

Not to say parallel arrangments do not have their place because they do. I am forced to do it all the time with 48 volt 40,000 AH telephone battery plants. I have to use parallel arrangements because about th elargest 2 volt battery that can be had is 4000 AH, so it takes 10 parallel strings to get 40,000 AH. Bu tI use extreme measures like a custom built over head copper buss deisnged to math raesistance and use a lot of special equipment to balance all the resistances well below 15 micro-ohms. That is not possible for Joe Home Owner or Alarm contractors.

Bottom line is batteries are available from just a few AH like Alarm and Fire Panels all the way up to 4000 AH in about any increment you need. It is just smart buisiness practice to use best practice and in this case is also the least expensive. If you are going to take this thing solar, you are going to need a lot larger battery than the alarm cabinet can hold and will have to have an external box mounted. Two 100 AH 12 volt batteries are more expensive than 1 200 AH 12 volt battery.

Once you figure out exactly what the requirements are, come back and I will help you select components.

[fraggedout]

Do no trun batteries in parallel, you are just asking for trouble. If you need 100 AH, then use a 100 AH battery

Thanks, Sunking, but please clarify why running batteries in parallel is more problematic than a single large battery. I ask this because the battery pigtails supplied by the manufacturer of the alarm panel have two sets of terminals, allowing for two 12VDC batteries of pick-your-AH-rating to be connected in parallel for longer reserve power. I do understand that if one of the batteries in a parallel setup goes bad, it will affect the other batteries as they will attempt to equalize across the array, but in my mind this will still keep the system somewhat functional until the bad battery in the array is replaced and max storage capacity is restored, whereas the failure of a single, large battery will render the system useless.

Please point out the flaw(s) in my logic.

[Sunking]

PS: I can also run several panel batteries in parallel to give myself much more battery capacity.

Do no trun batteries in parallel, you are just asking for trouble. If you need 100 AH, then use a 100 AH battery

[fraggedout]

PS: If you can figure out whether the AC is actually powering anything but the battery charger inside the alarm unit, you may find that you are able to fake it into thinking it is getting AC and so avoid the double conversion losses.

We've been using DSC alarm systems for years and there are programmable options that allow trouble conditions such as AC failure to be ignored (not displayed on the keypad with associated beeps), for turning off keypad backlighting, for blanking the keypad when not in use, etc., all of which would help conserve power. Further, while the detection devices to be used will be wireless (and thus use no power from the panel, having their own internal batteries), the panel will have to supply 12VDC power to the receiver for the wireless devices and the system's backup battery will have to supply continuous 12VDC operational power to the attached cellular alarm communicator, and that cellular communicator can draw up to 2 amps from the panel's backup battery (for a few seconds) when the system goes into alarm and the cellular transmitter communicates with Central Station. Sirens draw a lot of 12VDC power as well when the system goes into alarm, a large part of which is augmented by the panel's battery backup, but I can program the 40W siren to only sound for one minute rather than the default setting of four minutes.

I think I need to retrench a bit here and hook up a system at my shop with the configuration as it will be in the field (i.e., new fully charged battery without a 16.5VAC power supply) and do some testing. If it turns out that I have enough functionality running strictly off the system's 12VDC backup battery, then I see no reason to have to invert to 120VAC and then step back down to 16.5VAC.

PS: I can also run several panel batteries in parallel to give myself much more battery capacity. It may be simplistic thinking, in that I'm a noob, but what would prevent me from connecting the charge controller to a bank of 12VDC/12AH batteries in parallel and simply using the panel's existing 12VDC battery inputs? Assuming, of course, that I can retain enough functionality running off battery?

[inetdog]

Again, thanks to both of you for your prompt and patient replies, although I must correct Sunking in that 5th grade math elsewhere is post-doctoral study here in Arkansas. I'll take the advice and get a more accurate estimate of power drain at the panel and should be able to downsize the system (and the cost). Kill-a-watt meters are not very expensive and I'm certain I'll use it elsewhere. We have hundreds of alarm systems installed, so access to one (my home and my shop, for example) won't be a problem. Typically a new, fresh, fully-charged 12VDC 7AH SLA backup battery will power a panel in "backup" mode for about two days without AC input, so I should be able to figure the drain based on that. I'll revise my panel and battery calculations...

So the questions on "Are You Smarter Than a Fifth Grader" are considered really hard out there?

Just to jump start your calculations: The backup battery you describe holds 84 watt-hours of energy if you drain it to zero (not a good thing, but it does not happen very often in a real system).
If we figure 48 hours, that would be about 1.75 watts. It could use a whole lot more in full operation, and you will have some drain on the inverter's battery just to service the zero power idling current in the transformer. The Kill-a-Watt will tell you what that amounts to. Compare the Watt reading to the Volt-Amp (VA) reading. Even worse, the inverter itself will use a significant amount of power (a few watts at least) just to keep itself running. \

Looking forward to seeing the results.

PS: If you can figure out whether the AC is actually powering anything but the battery charger inside the alarm unit, you may find that you are able to fake it into thinking it is getting AC and so avoid the double conversion losses.

[fraggedout]

Good idea, but one which is amazingly uncommon among people who actually buy part of a system and then come here for help on the rest.
You came for help before you started, although only after getting your friend excited.


Yes. Two 12 volt 160 AH (by the way, it is AH, as in amps times hours rather than A/H as in amps per hour) batteries in series would be 12V, 320AH.
In parallel they would make 24V, 160AH. Four in series parallel (make two strings of two in series and then parallel the ends of the two strings. No connection across the middle) would give you 24V at 320AH for twice as much stored power as two batteries would give you.

If you go for FLA batteries instead of AGM, the cost will be about 1/2 as much. The first issue is to get a very accurate idea of what total power you need for the system.
By Sunking's estimate and calculation, you would only need two batteries, not 4, just using a 24 volt instead of a 12 volt inverter to get your 120 volt AC.

If the real power requirement of the alarm system is less than 30 watts, then you would also need less battery. You can determine the real power drain by using a Kill-a-Watt meter on the transformer input if you can get access to a working system OR by looking at how long the system is supposed to last on its internal battery and calculating the standby power drain. Then double it for a rough estimate of the AC operation power.

You also need to keep in mind that when the inverter shuts down from low battery, the system will still keep going for some time on its internal battery. Hopefully it will generate a trouble signal at that point so you could cart the batteries off to be recharged from AC and then bring them back.

Again, thanks to both of you for your prompt and patient replies, although I must correct Sunking in that 5th grade math elsewhere is post-doctoral study here in Arkansas. I'll take the advice and get a more accurate estimate of power drain at the panel and should be able to downsize the system (and the cost). Kill-a-watt meters are not very expensive and I'm certain I'll use it elsewhere. We have hundreds of alarm systems installed, so access to one (my home and my shop, for example) won't be a problem. Typically a new, fresh, fully-charged 12VDC 7AH SLA backup battery will power a panel in "backup" mode for about two days without AC input, so I should be able to figure the drain based on that. I'll revise my panel and battery calculations...

[Sunking]

I'm diagramming this out in my head as I go. If I run two 12VDC/160a/h AGM batteries in series, I get 24VDC @ 160a/h. If I run two 12VDC/160a/h AGMbatteries in parallel, I get 12VDC @ 320a/h. So should I run a parallel/series configuration to get 24VDC @ 320a/h? Did I even figure that right?

Yes, again 5th grade math.

AH = WH / V
3600 wh / 12 v = 300 ah = 3600 wh / 24 v = 150 ah.

So you buy 2-12 volt 150 AH batteries either way. or 1 single 12 volt 300 AH battery. Pretty easy puzzle to put together huh?

[inetdog]

I'm diagramming this out in my head as I go. If I run two 12VDC/160a/h AGM batteries in series, I get 24VDC @ 160a/h. If I run two 12VDC/160a/h AGMbatteries in parallel, I get 12VDC @ 320a/h. So should I run a parallel/series configuration to get 24VDC @ 320a/h? Did I even figure that right? And even hope I can remember how to wire a parallel/series configuration? Yikes, that's roughly $2500 worth of batteries, not to mention the cost of the panel, controller, inverter, cabling, and labor. My client may have a cow and decide to hire an armed guard. Ah, well, I'll run it by him and then duck. A lesson from the past remembered..."Engage brain before putting mouth in gear".

Good idea, but one which is amazingly uncommon among people who actually buy part of a system and then come here for help on the rest.
You came for help before you started, although only after getting your friend excited.


Yes. Two 12 volt 160 AH (by the way, it is AH, as in amps times hours rather than A/H as in amps per hour) batteries in series would be 12V, 320AH.
In parallel they would make 24V, 160AH. Four in series parallel (make two strings of two in series and then parallel the ends of the two strings. No connection across the middle) would give you 24V at 320AH for twice as much stored power as two batteries would give you.

If you go for FLA batteries instead of AGM, the cost will be about 1/2 as much. The first issue is to get a very accurate idea of what total power you need for the system.
By Sunking's estimate and calculation, you would only need two batteries, not 4, just using a 24 volt instead of a 12 volt inverter to get your 120 volt AC.

If the real power requirement of the alarm system is less than 30 watts, then you would also need less battery. You can determine the real power drain by using a Kill-a-Watt meter on the transformer input if you can get access to a working system OR by looking at how long the system is supposed to last on its internal battery and calculating the standby power drain. Then double it for a rough estimate of the AC operation power.

You also need to keep in mind that when the inverter shuts down from low battery, the system will still keep going for some time on its internal battery. Hopefully it will generate a trouble signal at that point so you could cart the batteries off to be recharged from AC and then bring them back.

[Sunking]

what are your opinions on 12VDC versus 24VDC and FLAs versus SLAs in an application such as mine?

For this application you are going to want AGM batteries. DO NOT confuse that with Gel batteries. They will last about 2 maybe 3 years, then you get a follow up call and more Jingle in your pocket for replacement.

[Sunking]

I think I just earned a gold star. Not that I would mistrust your calcs, sunking, but I needed to understand the calcs, not just take someone's word for it.

No problem. First calculation is straight forward energy use in a day. 30 watts x 24 hours = 720 watts hours. Just like miles per gallon in a car that gets 30 Mpg and you gotta go 720 miles. No matter how you spin it you will use 24 gallons.

The adjustment fact is to account for all the system losses in the wiring, charge controller, battery charge efficiency, and inverter conversion losses. All totaled up a Off-Grid Battery system with MPPT controller is 66%, and 50% for PWM. So if you need to use 720 watt hours, the panels have to make more than that. So at 66% efficiency to use 720 watts you got to make 720 / .66 = 1090 watthours.


With any battery system you have to use worse case which is winter. In your case 3.9 Hours right. Just simple 5th grade math to find the watts. WATTS = WATT HOUR / HOUR. 1090 wh / 3.9 h = 279 watts. Round that up to 300 watts and you are done with panel calculations.

Battery is easy fall off the log math. You need a battery with 5 days reserve capacity. So 5 x 720 wh = 3600 Watt Hours. . Now select battery voltage and you say 12 volt? Find the Amp Hours. Again simple 5th grade math, or 10 year college math in Arkansas. AH = WH / Battery Voltage. 3600 wh / 12 volts = 300 Amp Hours.

If using a MPPT charge controller the formula is Panel Wattage / Battery Voltage so 300 watts / 12 volts = 25 amps

All of it is very straight forward 5th grade math.

[fraggedout]

if going to 24 volts would allow you to use batteries in series instead of parallel, that would be a good thing all by itself. (300 AH at 12 volts is a large battery)
As to FLA versus AGM, the deciding factor may be whether you will be able to do the periodic maintenance (checking) on the batteries which FLA will need or have to have something (AGM) which is more maintenance free but costs twice as much and will require replacement twice as often. Avoid GEL batteries.
Sunking will probably also tell you that if you have limited solar hours during the day to recharge the batteries, AGM will be able to handle more panels (higher maximum charging rate) than the same capacity FLA batteries.

I'm diagramming this out in my head as I go. If I run two 12VDC/160a/h AGM batteries in series, I get 24VDC @ 160a/h. If I run two 12VDC/160a/h AGMbatteries in parallel, I get 12VDC @ 320a/h. So should I run a parallel/series configuration to get 24VDC @ 320a/h? Did I even figure that right? And even hope I can remember how to wire a parallel/series configuration? Yikes, that's roughly $2500 worth of batteries, not to mention the cost of the panel, controller, inverter, cabling, and labor. My client may have a cow and decide to hire an armed guard. Ah, well, I'll run it by him and then duck. A lesson from the past remembered..."Engage brain before putting mouth in gear". Seems like a whole lotta expense for a little old alarm system that runs off a 16.5VDC 40VA transformer.

[inetdog]

Thanks to both of you, sunking and inetdog. I read the posting you supplied the link for, sunking, and did the calculations myself and was pleased when my calcs matched yours. And inetdog, you understood what I was saying when I explained the need for inverting to 120VAC before stepping back down to 16.5VAC. So a big thanks to you both...

My location is Stuttgart, Arkansas, which is virtually the same latitude/elevation/weather as nearby Little Rock, Arkansas. According to http://www.solardirect.com/pv/system...sun-hours.html, Little Rock gets 5.29 hrs/avg summer sun, 3.88 hrs. winter sun, and a yearly average of 4.69. Assuming the worst-case scenario and figuring for winter months, the formula would then be:

30W/h x 24hrs. = 720 Watts/day
720 watts x 1.5 overhead = 1080 Watts/day
1080 watts/day divided by 3.88 Sun Hours (winter, Little Rock, Arkansas) = 278.35W, rounded up to a 400W panel

720W-H x 5 days = 3600W-H
3600W-H divided by 12V = 300A/h Battery

400W panel divided by 12V battery = 33.333A, rounded up to a 40A charge controller.

I think I just earned a gold star. Not that I would mistrust your calcs, sunking, but I needed to understand the calcs, not just take someone's word for it. Now, a couple of follow-up questions:
The author of the tutorial suggested never using 12V, opting for 24VDC with panels 300W or higher. He also cites pros and cons of using FLAs versus SLAs. In your experience, what are your opinions on 12VDC versus 24VDC and FLAs versus SLAs in an application such as mine?
Thanks again...

if going to 24 volts would allow you to use batteries in series instead of parallel, that would be a good thing all by itself. (300 AH at 12 volts is a large battery)
As to FLA versus AGM, the deciding factor may be whether you will be able to do the periodic maintenance (checking) on the batteries which FLA will need or have to have something (AGM) which is more maintenance free but costs twice as much and will require replacement twice as often. Avoid GEL batteries.
Sunking will probably also tell you that if you have limited solar hours during the day to recharge the batteries, AGM will be able to handle more panels (higher maximum charging rate) than the same capacity FLA batteries.

[fraggedout]

Thanks to both of you, sunking and inetdog. I read the posting you supplied the link for, sunking, and did the calculations myself and was pleased when my calcs matched yours. And inetdog, you understood what I was saying when I explained the need for inverting to 120VAC before stepping back down to 16.5VAC. So a big thanks to you both...

My location is Stuttgart, Arkansas, which is virtually the same latitude/elevation/weather as nearby Little Rock, Arkansas. According to http://www.solardirect.com/pv/system...sun-hours.html, Little Rock gets 5.29 hrs/avg summer sun, 3.88 hrs. winter sun, and a yearly average of 4.69. Assuming the worst-case scenario and figuring for winter months, the formula would then be:

30W/h x 24hrs. = 720 Watts/day
720 watts x 1.5 overhead = 1080 Watts/day
1080 watts/day divided by 3.88 Sun Hours (winter, Little Rock, Arkansas) = 278.35W, rounded up to a 400W panel

720W-H x 5 days = 3600W-H
3600W-H divided by 12V = 300A/h Battery

400W panel divided by 12V battery = 33.333A, rounded up to a 40A charge controller.

I think I just earned a gold star. Not that I would mistrust your calcs, sunking, but I needed to understand the calcs, not just take someone's word for it. Now, a couple of follow-up questions:
The author of the tutorial suggested never using 12V, opting for 24VDC with panels 300W or higher. He also cites pros and cons of using FLAs versus SLAs. In your experience, what are your opinions on 12VDC versus 24VDC and FLAs versus SLAs in an application such as mine?
Thanks again...